Molecular Characterization of GCP170, a 170-kDa Protein Associated with the Cytoplasmic Face of the Golgi Membrane
1997; Elsevier BV; Volume: 272; Issue: 38 Linguagem: Inglês
10.1074/jbc.272.38.23851
ISSN1083-351X
AutoresYoshio Misumi, Miwa Sohda, Akiko Yano, Toshiyuki Fujiwara, Yukio Ikehara,
Tópico(s)Endoplasmic Reticulum Stress and Disease
ResumoWe have isolated a cDNA clone encoding a protein (designated GCP170) of 1530 amino acid residues with a calculated molecular mass of 170 kDa that is localized to the Golgi complex. Hydropathy analysis shows that GCP170 contains no NH2-terminal signal sequence nor a hydrophobic domain sufficient for participating in membrane localization. It is also predicted that GCP170 has characteristic secondary structures including an extremely long α-helical domain that likely forms a coiled-coil between non-coil domains at the NH2 and COOH termini, suggesting that the protein is organized as a globular head, a stalk, and a tail. Immunocytochemical observations revealed that GCP170 was localized to the Golgi complex and the cytoplasm, consistent with biochemical data indicating that the protein exits as a membrane-associated form and a soluble form. GCP170 was dissociated from the Golgi membrane in response to brefeldin A as rapidly as a coat protein complex of non-clathrin-coated vesicles (β-COP, a subunit of coatomer), but did not co-localize with β-COP on the Golgi membrane when examined by immunoelectron microscopy. The protein was detected as phosphorylated and unphosphorylated forms, of which the unphosphorylated form was more tightly associated with the Golgi membrane. When cells were extracted with 1% Triton X-100 under microtubule-stabilizing conditions, GCP170 remained in the cells in association with the Golgi complex. These results indicate that GCP170 is a peripheral membrane protein with a long coiled-coil domain that may be involved in the structural organization or stabilization of the Golgi complex. We have isolated a cDNA clone encoding a protein (designated GCP170) of 1530 amino acid residues with a calculated molecular mass of 170 kDa that is localized to the Golgi complex. Hydropathy analysis shows that GCP170 contains no NH2-terminal signal sequence nor a hydrophobic domain sufficient for participating in membrane localization. It is also predicted that GCP170 has characteristic secondary structures including an extremely long α-helical domain that likely forms a coiled-coil between non-coil domains at the NH2 and COOH termini, suggesting that the protein is organized as a globular head, a stalk, and a tail. Immunocytochemical observations revealed that GCP170 was localized to the Golgi complex and the cytoplasm, consistent with biochemical data indicating that the protein exits as a membrane-associated form and a soluble form. GCP170 was dissociated from the Golgi membrane in response to brefeldin A as rapidly as a coat protein complex of non-clathrin-coated vesicles (β-COP, a subunit of coatomer), but did not co-localize with β-COP on the Golgi membrane when examined by immunoelectron microscopy. The protein was detected as phosphorylated and unphosphorylated forms, of which the unphosphorylated form was more tightly associated with the Golgi membrane. When cells were extracted with 1% Triton X-100 under microtubule-stabilizing conditions, GCP170 remained in the cells in association with the Golgi complex. These results indicate that GCP170 is a peripheral membrane protein with a long coiled-coil domain that may be involved in the structural organization or stabilization of the Golgi complex. The Golgi complex is a highly organized organelle comprised of thecis-Golgi network, Golgi stack, and trans-Golgi network and involved in transport, processing, and sorting of newly synthesized proteins including secretory, plasma membrane and lysosomal proteins (1Farquhar M.G. Annu. Rev. Cell Biol. 1985; 1: 447-488Crossref PubMed Scopus (395) Google Scholar, 2Rothman J.E. Orci L. Nature. 1992; 335: 409-415Crossref Scopus (744) Google Scholar). Various enzymes involved in proteolytic and oligosaccharide processing of the transported proteins have been shown to be localized to subcompartments of the Golgi complex (3Kornfeld R. Kornfeld S. Annu. Rev. Biochem. 1985; 54: 631-664Crossref PubMed Scopus (3776) Google Scholar, 4Paulson J.C. Colley K.J. J. Biol. Chem. 1989; 264: 17615-17618Abstract Full Text PDF PubMed Google Scholar, 5Misumi Y. Oda K. Fujiwara T. Takami N. Tashiro K. Ikehara Y. J. Biol. Chem. 1991; 266: 16954-16959Abstract Full Text PDF PubMed Google Scholar). These Golgi-resident enzymes are all transmembrane proteins with their major domains disposed to the lumen. In contrast, there are many cytosolic proteins and cytoplasmically disposed membrane proteins that are involved in the vesicular transport through the Golgi complex (2Rothman J.E. Orci L. Nature. 1992; 335: 409-415Crossref Scopus (744) Google Scholar, 6Rothman J.E. Nature. 1994; 372: 55-63Crossref PubMed Scopus (2007) Google Scholar). Coatomer (COP) 1The abbreviations used are: COP, coat protein complex of non-clathrin-coated vesicles; BFA, brefeldin A; GFP, green fluorescent protein; Man II, α-mannosidase II; PAGE, polyacrylamide gel electrophoresis; SNAP, solubleN-ethylmaleimide-sensitive fusion protein attachment protein; SNARE, SNAP receptor; PCR, polymerase chain reaction; kb, kilobase(s); PIPES, 1,4-piperazinediethanesulfonic acid; GTPγS, guanosine 5′-3-O-(thio)triphosphate. and ADP-ribosylation factor are required for budding of transport vesicles from the Golgi membrane (7Waters M.G. Serafini T. Rothman J.E. Nature. 1991; 349: 248-251Crossref PubMed Scopus (379) Google Scholar, 8Serafini T. Orci L. Amherdt M. Brunner M. Kahn R.A. Rothman J.E. Cell. 1991; 67: 239-253Abstract Full Text PDF PubMed Scopus (451) Google Scholar, 9Orci L. Palmer D.J. Amherdt M. Rothman J.E. Nature. 1993; 364: 732-734Crossref PubMed Scopus (182) Google Scholar). Other cytosolic proteins such asN-ethylmaleimide-sensitive fusion protein and solubleN-ethylmaleimide-sensitive fusion protein attachment proteins (SNAPs) and membrane-associated SNAP receptors (SNAREs) are required for attachment and/or fusion of the vesicles to target membranes (10Malhotra V. Orci L. Glick B.S. Block M.R. Rothman J.E. Cell. 1988; 54: 221-227Abstract Full Text PDF PubMed Scopus (266) Google Scholar, 11Clary D.O. Griff I.C. Rothman J.E. Cell. 1990; 61: 709-721Abstract Full Text PDF PubMed Scopus (404) Google Scholar, 12Söllner T. Whiteheart S.W. Brunner M. Erdjument-Bromage H. Geromanos S. Tempst p. Rothman J.E. Nature. 1993; 362: 318-324Crossref PubMed Scopus (2628) Google Scholar), leading to the proposal of the SNARE hypothesis (6Rothman J.E. Nature. 1994; 372: 55-63Crossref PubMed Scopus (2007) Google Scholar,12Söllner T. Whiteheart S.W. Brunner M. Erdjument-Bromage H. Geromanos S. Tempst p. Rothman J.E. Nature. 1993; 362: 318-324Crossref PubMed Scopus (2628) Google Scholar). The Golgi complex has morphological characteristics that include ordered (from cis to trans) stacking, close apposition, and constant spacing of stacked cisternae. The characteristic stacked structure of the Golgi complex is rapidly disrupted by treatment of cells with brefeldin A (BFA) (13Misumi Y. Misumi Y. Miki K. Takatsuki A. Tamura G. Ikehara Y. J. Biol. Chem. 1986; 261: 11398-11403Abstract Full Text PDF PubMed Google Scholar, 14Fujiwara T. Oda K. Yokota S. Takatsuki A. Ikehara Y. J. Biol. Chem. 1988; 263: 18545-18552Abstract Full Text PDF PubMed Google Scholar, 15Fujiwara T. Oda K. Ikehara Y. Cell Struct. Funct. 1989; 14: 605-616Crossref PubMed Scopus (35) Google Scholar, 16Lippincott-Schwartz J. Yuan L.C. Bonifacino J.S. Klausner R.D. Cell. 1989; 56: 801-813Abstract Full Text PDF PubMed Scopus (1312) Google Scholar). The evidence that BFA primarily blocks the budding of transport vesicles (6Rothman J.E. Nature. 1994; 372: 55-63Crossref PubMed Scopus (2007) Google Scholar, 17Donaldson J.G. Finazzi D. Klausner R.D. Nature. 1992; 360: 350-352Crossref PubMed Scopus (595) Google Scholar, 18Helms J.B. Rothman J.E. Nature. 1992; 360: 352-354Crossref PubMed Scopus (581) Google Scholar) indicates that the maintenance of the Golgi structure is supported by the vesicular transport system. In addition, it has been suggested that the stacking of Golgi cisternae can be explained by an extension of the SNARE hypothesis (19Rothman J.E. Warren G. Curr. Biol. 1994; 4: 220-233Abstract Full Text Full Text PDF PubMed Scopus (481) Google Scholar); stacking is most simply viewed as an extension of the docking process with the heterotypic v-SNAREs in each cisterna interacting with the cognate t-SNAREs in the next. This would require a mechanism preventing the fusion that leads to vesicular transport, postulating the existence of "fusion clamps" (19Rothman J.E. Warren G. Curr. Biol. 1994; 4: 220-233Abstract Full Text Full Text PDF PubMed Scopus (481) Google Scholar). Whether or not the SNARE system is involved in this process, there may be cytosolic and/or cytoplasmically disposed membrane proteins responsible for the structural maintenance of the Golgi complex, although little is known about them. Recently several new Golgi-associated proteins have been identified by antibodies from patients with autoimmune diseases and by antibodies prepared in animals. These include giantin or GCP372 (20Linstedt A.D. Hauri H.-P. Mol. Biol. Cell. 1993; 4: 679-693Crossref PubMed Scopus (355) Google Scholar, 21Sohda M. Misumi Y. Fujiwara T. Nishioka M. Ikehara Y. Biochem. Biophys. Res. Commun. 1994; 205: 1399-1408Crossref PubMed Scopus (42) Google Scholar, 22Seelig H.P. Schranz P. Schroter H. Wiemann C. Renz M. J. Autoimmun. 1994; 7: 67-91Crossref PubMed Scopus (60) Google Scholar), golgin-245 or p230 (23Fritzler M.J. Lung C.-C. Hamel J.C. Griffith J. Chan E.K.L. J. Biol. Chem. 1995; 270: 31262-31268Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar, 24Erlich R. Gleeson P.A. Campbell P. Dietzsch E. Toh B.-H. J. Biol. Chem. 1996; 271: 8328-8337Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar), GCP230 (25Funaki T. Fujiwara T. Hong H.-S. Misumi Y. Nishioka M. Ikehara Y. Cell Struct. Funct. 1996; 21: 63-72Crossref PubMed Scopus (13) Google Scholar), p210 (26Rios R.M. Tassin A.-M. Celati C. Antony C. Boissier M.-C. Hamberg J.-C. Bornens M. J. Cell Biol. 1994; 125: 997-1013Crossref PubMed Scopus (68) Google Scholar), and GM130 or golgin-95 (27Nakamura N. Rabouille C. Watson R. Nilsson T. Hui N. Slusarewicz P. Kreis T.E. Warren G. J. Cell Biol. 1995; 131: 1715-1726Crossref PubMed Scopus (676) Google Scholar, 28Fritzler M.J. Hamel J.C. Ochs R.L. Chan E.K.L. J. Exp. Med. 1993; 178: 49-62Crossref PubMed Scopus (135) Google Scholar). These proteins have several characteristic features. Their molecular masses are quite large, ranging from 370 kDa for the largest giantin/GCP372 to 130 kDa for the smallest GM130/golgin-95. All of them are associated with the cytoplasmic face of the Golgi membrane. Although giantin/GCP372 is an integral membrane protein anchored to the membrane by the COOH-terminal hydrophobic domain (21Sohda M. Misumi Y. Fujiwara T. Nishioka M. Ikehara Y. Biochem. Biophys. Res. Commun. 1994; 205: 1399-1408Crossref PubMed Scopus (42) Google Scholar, 22Seelig H.P. Schranz P. Schroter H. Wiemann C. Renz M. J. Autoimmun. 1994; 7: 67-91Crossref PubMed Scopus (60) Google Scholar), all the other proteins have no hydrophobic domain that could function as a signal sequence or participate in membrane localization. The most characteristic feature is that all of the proteins have extensively large domains, enabling the formation of coiled-coil structures analogous to the myosin family (21Sohda M. Misumi Y. Fujiwara T. Nishioka M. Ikehara Y. Biochem. Biophys. Res. Commun. 1994; 205: 1399-1408Crossref PubMed Scopus (42) Google Scholar, 22Seelig H.P. Schranz P. Schroter H. Wiemann C. Renz M. J. Autoimmun. 1994; 7: 67-91Crossref PubMed Scopus (60) Google Scholar, 23Fritzler M.J. Lung C.-C. Hamel J.C. Griffith J. Chan E.K.L. J. Biol. Chem. 1995; 270: 31262-31268Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar, 24Erlich R. Gleeson P.A. Campbell P. Dietzsch E. Toh B.-H. J. Biol. Chem. 1996; 271: 8328-8337Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar, 25Funaki T. Fujiwara T. Hong H.-S. Misumi Y. Nishioka M. Ikehara Y. Cell Struct. Funct. 1996; 21: 63-72Crossref PubMed Scopus (13) Google Scholar, 26Rios R.M. Tassin A.-M. Celati C. Antony C. Boissier M.-C. Hamberg J.-C. Bornens M. J. Cell Biol. 1994; 125: 997-1013Crossref PubMed Scopus (68) Google Scholar, 27Nakamura N. Rabouille C. Watson R. Nilsson T. Hui N. Slusarewicz P. Kreis T.E. Warren G. J. Cell Biol. 1995; 131: 1715-1726Crossref PubMed Scopus (676) Google Scholar, 28Fritzler M.J. Hamel J.C. Ochs R.L. Chan E.K.L. J. Exp. Med. 1993; 178: 49-62Crossref PubMed Scopus (135) Google Scholar). Several other proteins involved in membrane targeting and fusion also contain long coiled-coil domains, which are implicated in interactions leading to vesicle targeting and/or fusion (29Achstetter T. Franzusoff A. Field C. Schekman R. J. Biol. Chem. 1988; 263: 11711-11717Abstract Full Text PDF PubMed Google Scholar, 30Südhof T.C. Czernik A.J. Kao H.-T. Takei K. Johnston P.A. Horiuchi A. Kanazir S.D. Wagner M.A. Perin M.S. Camilli P.D. Greengard P. Science. 1989; 245: 1474-1480Crossref PubMed Scopus (421) Google Scholar, 31Sapperstein S.K. Watter D.M. Grosvenor A.R. Heuser J.E. Waters G. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 522-526Crossref PubMed Scopus (174) Google Scholar). However, the physiological function of the above mentioned proteins remains to be determined. Golgin-160 is another Golgi-associated protein that was also identified by human autoantibodies (28Fritzler M.J. Hamel J.C. Ochs R.L. Chan E.K.L. J. Exp. Med. 1993; 178: 49-62Crossref PubMed Scopus (135) Google Scholar), but its structure and properties have not been well characterized. In the present study, we report the use of a partial cDNA sequence from golgin-160 to isolate a full-length cDNA clone encoding a 170-kDa protein (termed GCP170) and describe the properties of this protein in comparison with those of other Golgi-associated proteins. Western blotting detection kit was purchased from Amersham Corp. (Tokyo, Japan). Anti-α-tubulin IgG was from Funakoshi (Tokyo, Japan); rhodamine-conjugated goat anti-rabbit IgG was from DAKO Japan (Tokyo, Japan); and fluorescein isothiocyanate-conjugated goat anti-guinea pig IgG was from Cappel Laboratories (West Chester, PA). Guinea pig anti-human GCP372 and rabbit anti-β-COP antibodies were prepared as described previously (21Sohda M. Misumi Y. Fujiwara T. Nishioka M. Ikehara Y. Biochem. Biophys. Res. Commun. 1994; 205: 1399-1408Crossref PubMed Scopus (42) Google Scholar, 25Funaki T. Fujiwara T. Hong H.-S. Misumi Y. Nishioka M. Ikehara Y. Cell Struct. Funct. 1996; 21: 63-72Crossref PubMed Scopus (13) Google Scholar). Rabbit anti-α-mannosidase II (Man II) was supplied by Dr. K. W. Moremen (University of Georgia). HepG2 (human hepatocarcinoma), QGP-1 (human pancreas carcinoma), HeLa (human cervix carcinoma), COS-1 (monkey kidney), and baby hamster kidney cells were cultured as described (21Sohda M. Misumi Y. Fujiwara T. Nishioka M. Ikehara Y. Biochem. Biophys. Res. Commun. 1994; 205: 1399-1408Crossref PubMed Scopus (42) Google Scholar, 25Funaki T. Fujiwara T. Hong H.-S. Misumi Y. Nishioka M. Ikehara Y. Cell Struct. Funct. 1996; 21: 63-72Crossref PubMed Scopus (13) Google Scholar). Poly(A)+ RNA prepared from QGP-1 cells was used for construction of a cDNA library in λZAPII bacteriophage. A golgin-160 cDNA fragment (28Fritzler M.J. Hamel J.C. Ochs R.L. Chan E.K.L. J. Exp. Med. 1993; 178: 49-62Crossref PubMed Scopus (135) Google Scholar) was prepared by reverse transcription-PCR from QGP-1 RNA to obtain a probe for screening and Northern blot analysis. Primers used in the PCR reaction were 5′-GGATCCTGCAGCAGGAGAC-3′ and 5′-GCTTGGAACTGTGCTATCTC-3′. Screening of 1 × 106 independent clones yielded 11 positive clones. These clones were subcloned into pBluescript SK− plasmid vector by automatic excision process (32Misumi Y. Sohda M. Okubo K. Takami N. Oda K. Ikehara Y. J. Biochem. ( Tokyo ). 1990; 108: 230-234Crossref PubMed Scopus (20) Google Scholar). A cDNA clone with the longest insert (3.5 kilobase pairs, named QSY103) was subcloned and subjected to nucleotide sequence determination by the dideoxynucleotide chain termination method (33Sanger F. Nicklen S. Coulson A.R. Proc. Natl. Acad. Sci. U. S. A. 1977; 74: 5463-5467Crossref PubMed Scopus (52678) Google Scholar). To obtain a full-length cDNA, we constructed another cDNA library using a poly(A)+ RNA fraction that was enriched with 7-kb RNA hybridized with a fragment of QSY103 (34Misumi Y. Tashiro K. Hattori M. Sakaki Y. Ikehara Y. Biochem. J. 1988; 249: 661-668Crossref PubMed Scopus (44) Google Scholar). Screening 5 × 105 independent clones of the new cDNA library yielded 12 positive clones. A clone with the longest insert obtained (FQSY1024; see Fig. 1) was subcloned, and the nucleotide sequences were determined as above. Site-directed mutagenesis was carried out to introduce a new EcoRI site at nucleotide 260 and to destroy theEcoRI site at nucleotide 1039 in the GCP170 cDNA (FQSY1024), yielding a clone FQSY1024m that encodes the same amino acid sequence of GCP170. To construct an expression plasmid for fluorescence-microscopic observations, anEcoRI-EcoRI fragment (6.4 kilobase pairs) of FQSY1024m was inserted into the EcoRI site of pGFPC1 (pGFPC1/GCP170), which expressed GCP170 as a fusion protein with the green fluorescent protein (GFP). For metabolic labeling experiments, the EcoRI-EcoRI fragment of FQSY1024m was inserted into the EcoRI site of pSG5 expression vector (pSG/GCP170). Each purified plasmid (10 μg) was transfected into baby hamster kidney cells or COS-1 cells as described previously (5Misumi Y. Oda K. Fujiwara T. Takami N. Tashiro K. Ikehara Y. J. Biol. Chem. 1991; 266: 16954-16959Abstract Full Text PDF PubMed Google Scholar). Chimeric proteins of GCP170 fragments fused to the COOH terminus of glutathioneS-transferase were prepared using the expression vector pGEX3X (35Smith D.B. Johnson K.S. Gene ( Amst .). 1988; 67: 31-40Crossref PubMed Scopus (5047) Google Scholar). Two cDNA fragments of GCP170 (corresponding to amino acid number 827–990 and 1136–1231 in Fig. 2) prepared in pT7-blue were digested with BamHI and EcoRI, purified by gel electrophoresis, and ligated to theBamHI-EcoRI site of pGEX3X vector. The glutathione S-transferase fusion proteins with expected molecular masses were obtained by culture of recombinant bacteria and purified (35Smith D.B. Johnson K.S. Gene ( Amst .). 1988; 67: 31-40Crossref PubMed Scopus (5047) Google Scholar). These two recombinant proteins were mixed and injected into rabbits to raise anti-GCP170 antibodies. Immunofluorescence and immunoelectron microscopic observations were carried out as described before (21Sohda M. Misumi Y. Fujiwara T. Nishioka M. Ikehara Y. Biochem. Biophys. Res. Commun. 1994; 205: 1399-1408Crossref PubMed Scopus (42) Google Scholar). Primary antibodies used for immunofluorescence staining were as follows; rabbit anti-GCP170 serum (dilution factor, 1:50), anti-α-tubulin serum (1:100), anti-β-COP IgG (15 μg/ml), and guinea pig anti-GCP372 serum (1:50). Rhodamine-conjugated goat anti-rabbit IgG (1:50) or fluorescein isothiocyanate-conjugated anti-guinea pig IgG (1:50) was used as a secondary antibody. When indicated, cells were extracted with 1% Triton X-100, 80 mm PIPES buffer (pH 6.8), 1 mmMgCl2, 2 mm EGTA, and 2 m glycerol (microtubule-stabilizing buffer), or with 1% Triton X-100, 10 mm phosphate buffer (pH 7.2), 5 mmCaCl2, and 150 mm NaCl (microtubule-destabilizing buffer) before fixation. Cells were also treated with BFA or with nocodazole as described (25Funaki T. Fujiwara T. Hong H.-S. Misumi Y. Nishioka M. Ikehara Y. Cell Struct. Funct. 1996; 21: 63-72Crossref PubMed Scopus (13) Google Scholar, 36Turner J.R. Tartakoff A.M. J. Cell Biol. 1989; 109: 2081-2088Crossref PubMed Scopus (138) Google Scholar). For immunoelectron microscopic observations, rabbit anti-GCP170 serum (1:100) or anti-β-COP IgG (10 μg/ml) was used as a primary antibody, which was detected with gold-conjugated protein A. This was performed as described previously (25Funaki T. Fujiwara T. Hong H.-S. Misumi Y. Nishioka M. Ikehara Y. Cell Struct. Funct. 1996; 21: 63-72Crossref PubMed Scopus (13) Google Scholar). In brief, proteins separated by SDS-PAGE (5 or 7% gels) were transferred onto a polyvinylidene difluoride membrane (Millipore), followed by incubation with anti-GCP170 serum (1:800) or with anti-Mann II serum (1:1000). Peroxidase-conjugated anti-rabbit IgG antibodies (1:2000) were used as secondary antibodies. The immunoreactive proteins were visualized using the Enhanced Chemiluminescence (ECL) kit. HepG2 cells were homogenized in 0.25 m sucrose, 10 mm Tris-HCl (pH 7.5), and 2 mm EDTA (homogenizing buffer) with a nitrogen bombardment apparatus (Parr Instrument Co.), followed by centrifugation at 1,000 × g for 10 min. The resultant postnuclear supernatant was separated by centrifugation at 105,000 × g for 1 h into a pellet (membrane fraction) and a supernatant (cytosol fraction). A Golgi fraction was prepared by flotation of the postnuclear supernatant in a sucrose gradient as described by Balch et al. (37Balch W.E. Dunphy W.G. Praell W.A. Rothman J.E. Cell. 1984; 39: 405-416Abstract Full Text PDF PubMed Scopus (480) Google Scholar). The Golgi membrane was sedimentated by centrifugation at 105,000 × g for 30 min and resuspended in the homogenizing buffer containing either 0–0.5m KCl, 1 m NaCl, or 0.1 mNa2CO3. After being incubated on ice for 30 min, the suspensions were centrifuged at 105,000 × gfor 30 min to obtain supernatants and membrane pellets. Phase separation of the indicated samples in a Triton X-114 solution was carried out by the method of Bordier (38Bordier C. J. Biol. Chem. 1981; 256: 1604-1607Abstract Full Text PDF PubMed Google Scholar). A Golgi fraction (1 mg protein/ml) was incubated at 0 °C for 30 min with a mixture of chymotrypsin and trypsin (50 μg each/ml) in 10 mm Tris-HCl (pH 8.0) and 150 mm NaCl with or without 1% Triton X-100. HepG2 cells (4 × 106 cells/dish) or transfected COS-1 cells (5 × 106 cells/dish) were labeled at 37 °C for 5 h with [35S]methionine (4 MBq/dish) in 2 ml of minimum essential medium lacking unlabeled methionine or with [32P]orthophosphate (20 MBq/dish) in 2 ml of phosphate-free minimum essential medium. Cell lysates were prepared in a solution containing 1% Triton X-100, 1% sodium deoxycholate, 0.01% SDS, and a protease inhibitor mixture and subjected to immunoprecipitation with anti-GCP170 antibodies in combination with protein A-Sepharose as described previously (13Misumi Y. Misumi Y. Miki K. Takatsuki A. Tamura G. Ikehara Y. J. Biol. Chem. 1986; 261: 11398-11403Abstract Full Text PDF PubMed Google Scholar). The immunocomplexes were boiled in Laemmli's sample buffer and analyzed by SDS-PAGE (5% or 7% gels) and fluorography (39Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207233) Google Scholar). The immunoprecipitates prepared from 35S-labeled cells were suspended in 50 mm Tris-HCl (pH 9.0) and incubated at 37 °C for 30 min in the presence or the absence of bovine intestinal alkaline phosphatase (50 units/ml), followed by SDS-PAGE and fluorography as above. Nucleotide and protein sequences were analyzed by the GENETX genetic information processing software (Software Development Co., Tokyo, Japan) and FASTA program at the DNA Data Bank of Japan (Mishima, Japan). Coiled-coil graphs were made from data generated by the COILS (version 2.1) algorithm of Swiss Cancer Research Institute (40Lupas A. van Dyke M. Stock J. Science. 1991; 252: 1162-1164Crossref PubMed Scopus (3471) Google Scholar). The MTDIK matrix and a window size of 28 residues were used. Based on the available SY2 sequence (28Fritzler M.J. Hamel J.C. Ochs R.L. Chan E.K.L. J. Exp. Med. 1993; 178: 49-62Crossref PubMed Scopus (135) Google Scholar), a cDNA fragment was obtained by the reverse transcription-PCR method using QGP-1 RNA as a template (PCR probe in Fig. 1 A). Screening of 1 × 106 clones of the QGP-1 cDNA library with the probe yielded 11 positive clones, of which clone QSY103 had the longest insert of 3.3 kb. Because Northern blot analysis showed the presence of a single 7-kb RNA (Fig. 1 B), we further screened another cDNA library constructed from a 7-kb RNA-enriched fraction and finally isolated a clone, FQSY1024, with a longer insert of 6.6 kb containing an open reading frame (Fig. 1 A). The open reading frame encodes a protein of 1530 amino acid residues with a calculated mass of 170 kDa (Fig.2 A). We designated the protein GCP170 (for the Golgi complex-associatedprotein of 170 kDa). The protein does not contain a sequence characteristic of a signal sequence at the NH2 terminus nor a hydrophobic domain sufficient to span a membrane in the entire sequence, as shown by a hydropathy analysis (Fig. 2 B). Search of the protein data base shows that GCP170 has no overall homology to known proteins nor particular motifs except for phosphorylation sites. Secondary structure analysis for coiled-coil probabilities (40Lupas A. van Dyke M. Stock J. Science. 1991; 252: 1162-1164Crossref PubMed Scopus (3471) Google Scholar) predicts that GCP170 consists of three major domains: an NH2-terminal non-coil (positions 1–400), a long α-helical domain with strong coiled-coil forming potential (positions 400–1400), and a COOH-terminal non-coil domain (1400–1530), although the coiled-coil domain has an interruption (about 60 residues) (Fig. 2 C). The NH2-terminal non-coil domain contains a proline-rich region and a serine-rich region (Fig. 2 A). Thus, it is likely that GCP170 has a domain structure with a globular head, long stalk, and short tail, similar to kinesin (41Hirokawa N. Trends Cell Biol. 1996; 6: 135-141Abstract Full Text PDF PubMed Scopus (132) Google Scholar) and CLIP-170 (42Rickard J.E. Kreis T.E. Trends Cell Biol. 1996; 6: 178-183Abstract Full Text PDF PubMed Scopus (86) Google Scholar), although they have no significant homology to each other. Immunofluorescence microscopy with anti-recombinant GCP170 showed that GCP170 was concentrated at juxta-nuclear regions corresponding to the Golgi complex, although its presence in the cytoplasm was also suggested by faint and diffuse staining of whole cells (Fig. 3 a). The same perinuclear regions were co-stained with antibodies to GCP372, a Golgi-membrane-anchored protein (21Sohda M. Misumi Y. Fujiwara T. Nishioka M. Ikehara Y. Biochem. Biophys. Res. Commun. 1994; 205: 1399-1408Crossref PubMed Scopus (42) Google Scholar) (Fig. 3 b). The general staining pattern of GCP170 was most similar to that of β-COP, a subunit of coatomer (Fig. 3 c). Both GCP170 and β-COP displayed heavy staining of the Golgi region and faint staining throughout the cytoplasm, in contrast to the staining profile of GCP372, which was confined to the Golgi region. A more detailed localization of GCP170 was attained by immunoelectron microscopy (Fig.3 d). The immunogold particles were detected at the rims of cisternal structures and related elements characteristic of the Golgi complex. The gold particles were not significantly detected in other organelles including nuclei, mitochondria, and lysosomes, although some were found in the cytoplasm free from membrane association. A postnuclear fraction of HepG2 cells was subfractionated into cytosol, total membrane, and Golgi fractions, which were analyzed by Western blotting. Anti-GCP170 antibodies recognized a 170-kDa protein not only in the Golgi and total membrane fractions but also in the cytosol (Fig.4 A). GCP170 was found to be more abundant in the cytosol than in the total membranes, although it was much enriched in the Golgi fraction. The membrane topology of GCP170 was examined using protease treatment. When the intact Golgi fraction was incubated with trypsin and chymotrypsin, GCP170 was completely digested even in the absence of a detergent (Fig.4 B, upper panel), whereas Man II, which has a cytoplasmic tail of only 5 amino acids, was degraded only when incubated with the detergent (Fig. 4 B, lower panel). In addition, GCP170 was completely released from the membrane with sodium carbonate, although not completely extracted with 1 m NaCl, and entirely partitioned into an aqueous phase when subjected to Triton X-114 phase separation (Fig. 4 C). Taken together, these results indicate that GCP-170 is a peripheral protein associated with the cytoplasmic face of the Golgi membrane. The product encoded by the GCP170 cDNA was examined by transfection experiments in baby hamster kidney cells. Because the anti-GCP170 antibodies were found to cross-react with cells derived from other species and could not distinguish exogenously introduced GCP170 from the endogenous protein, we transfected the plasmid pGFPC1/GCP170 into cells, which expressed a fusion protein containing the GFP as a reporter. Fluorescence microscopy shows essentially the same profile between the immunostaining and the GFP image (Fig.5 A, a andb), demonstrating that the fusion protein expressed is concentrated in the Golgi complex. Immunoprecipitation experiments were also carried out, for which cells were transfected with the plasmid pSG5 carrying the GCP170 cDNA and metabolically labeled with [35S]methionine. GCP170 expressed in the transfected cells was very heavily labeled and had the same molecular mass as that from HepG2 cells (Fig. 5 B, lanes 1 and 3). In this experiment, we used 5% gels for SDS-PAGE, under the conditions of which GCP170 was resolved into two closely migrating forms, although they migrated as an apparently single form when analyzed by SDS-PAGE on 7% gels (Fig. 4). Smaller bands detected in the transfected cells (Fig. 5 B, lane 1) may be mostly degradation products of GCP170. The presence of the doublet GCP170 (Fig. 5 B) may be explained by the expression of independent molecules with similar epitopes reactive with the anti-GCP170 or by modifications such as phospho
Referência(s)